Helix formation and capping energetics of arginine analogs with varying side chain length

Article abstract of DOI:10.1007/s00726-011-1064-2

Arginine (Arg) has been used for recognizing negatively charged biological molecules, cell penetration, and oligosaccharide mass signal enhancement. The versatility of Arg has inspired the need to develop Arg analogs and to research the structural effects of incorporating Arg analogs. Accordingly, we investigated the effect of Arg side chain length on helix formation by studying 12 Ala-based peptides containing the Arg analogs (S)-2-amino-6-guanidino- hexanoic acid (Agh), (S)-2-amino-4-guanidinobutyric acid (Agb), and (S)-2-amino-3-guanidinopropionic acid (Agp). Solid phase guanidinylation with orthogonal protection strategies was necessary to synthesize Agb- and Agp-containing peptides using Fmoc-based chemistry. The fraction helix for the peptides was determined by circular dichroism spectroscopy, and used to derive the statistical mechanical parameters and energetics for N-capping, C-capping, and helix propagation (propensity). All four Arg analogs were unfavorable for N-capping. The C-cap parameter followed the trend Agp < Agb < Arg < Agh, showing more favorable C-cap energetics with increasing side chain length. In contrast, helix propensity followed the trend Agp < Agb < Arg > Agh, highlighting the uniqueness of the Arg side chain length in helix formation. Molecular mechanics calculations and a survey on protein structures were consistent with the experimental results. Furthermore, calculations and survey both showed that the g- conformation for the χ₁ dihedral was present for the first two residues at the N-terminus of helices, but not favored in the center or C-terminus of helices due to sterics. These results should serve as the foundation for developing Arg-related bioactive compounds and technologies.

Full text of DOI:10.1007/s00726-011-1064-2

Amino Acids (2012) 43:195–206
DOI 10.1007/s00726-011-1064-2
ORIGINAL ARTICLE
Helix formation and capping energetics of arginine analogs
with varying side chain length
•
Richard P. Cheng Yi-Jen Weng Wei-Ren Wang Marc J. Koyack
•
•
•
•
Yuta Suzuki Cheng-Hsun Wu Po-An Yang Hao-Chun Hsu
•
•
•
•
•
Hsiou-Ting Kuo Prashant Girinath Chun-Jen Fang
Received: 12 June 2011 / Accepted: 20 August 2011 / Published online: 16 September 2011
ꢀ Springer-Verlag 2011
Abstract Arginine (Arg) has been used for recognizing
negatively charged biological molecules, cell penetration,
and oligosaccharide mass signal enhancement. The versa-
tility of Arg has inspired the need to develop Arg analogs
and to research the structural effects of incorporating Arg
analogs. Accordingly, we investigated the effect of Arg side
chain length on helix formation by studying 12 Ala-based
peptides containing the Arg analogs (S)-2-amino-6-guani-
dino-hexanoic acid (Agh), (S)-2-amino-4-guanidinobutyric
acid (Agb), and (S)-2-amino-3-guanidinopropionic acid
(Agp). Solid phase guanidinylation with orthogonal pro-
tection strategies was necessary to synthesize Agb- and
Agp-containing peptides using Fmoc-based chemistry. The
fraction helix for the peptides was determined by circular
dichroism spectroscopy, and used to derive the statistical
mechanical parameters and energetics for N-capping,
C-capping, and helix propagation (propensity). All four
Arg analogs were unfavorable for N-capping. The C-cap
parameter followed the trend Agp \ Agb \ Arg \ Agh,
showing more favorable C-cap energetics with increasing
side chain length. In contrast, helix propensity followed the
trend Agp \ Agb \ Arg [ Agh, highlighting the unique-
ness of the Arg side chain length in helix formation.
Molecular mechanics calculations and a survey on protein
structures were consistent with the experimental results.
Furthermore, calculations and survey both showed that the
g– conformation for the v1 dihedral was present for the first
two residues at the N-terminus of helices, but not favored in
the center or C-terminus of helices due to sterics. These
results should serve as the foundation for developing
Arg-related bioactive compounds and technologies.
Keywords Arginine ꢀ Side chain length ꢀ Peptide ꢀ Helix
Introduction
Arginine and lysine are the two positively charged amino
acids with side chain pKa values greater than ten (Fig. 1).
Incorporated into proteins, Arg is important for recognizing
anionic biological entities such as phosphates (Hirsch et al.
2007), glycosaminoglycans (Blaum et al. 2010), and others
(Chakrabarti 1994). Furthermore, Arg is responsible for the
cell penetration activity of Arg-containing proteins/pep-
tides (Wender et al. 2000, 2008; Mitchell et al. 2000), and
is frequently observed at protein–protein interaction inter-
faces and hotspots (Janin et al. 1988; Argos 1988; Jones
and Thornton 1995; Tsai et al. 1997; Bogan and Thorn
1998; Hu et al. 2000). For technological advantages, Arg
derivatives have been covalently attached to carbohydrates
to enhance the detection sensitivity of oligosaccharides in
MALDI (matrix-assisted laser desorption ionization) mass
spectrometry analysis (Zhao et al. 1997; Shinohara et al.
2004; Baumgart et al. 2004; Nishikaze and Takayama
2006; Northern et al. 2008). Arg is different from Lys in
bearing a guanidinium group to carry the positive charge
(Fig. 1). This creates a more diffuse positive charge com-
pared to the Lys ammonium group. Furthermore, the
Electronic supplementary material The online version of this
article (doi:10.1007/s00726-011-1064-2) contains supplementary
material, which is available to authorized users.
R. P. Cheng (&) ꢀ Y.-J. Weng ꢀ W.-R. Wang ꢀ C.-H. Wu ꢀ
P.-A. Yang ꢀ H.-C. Hsu ꢀ H.-T. Kuo ꢀ C.-J. Fang
Department of Chemistry, National Taiwan University,
Taipei 10617, Taiwan
e-mail: rpcheng@ntu.edu.tw
M. J. Koyack ꢀ Y. Suzuki ꢀ P. Girinath
Department of Chemistry, University at Buffalo, The State
University of New York, Buffalo, NY 14260-3000, USA
123

196
R. P. Cheng et al.
guanidinium group is capable of forming multiple hydro-
gen bonds in a multidentate fashion. This unique guanidi-
nium bearing side chain of Arg is critical for various
bioactivities such as cell penetration (Mitchell et al. 2000),
and cannot be replaced with Lys. Cell penetration activity
can be achieved by attaching the guanidinium group onto
the side chain of oligomers with non-natural backbones
including peptoids (Wender et al. 2000), oligo-carbamates
(Wender et al. 2002), and b-peptides (Umezawa et al.
2001). Accordingly, it is important to synthesize guanidi-
nium-containing Arg analogs and to investigate the struc-
tural and functional effects of these analogs.
and Kennan 2008), and Agh-containing 14-mers (Zhang
and Kennan 2001), whereas protocols using Fmoc-based
chemistry remain to be fully demonstrated.
The ability to generate Arg analog-containing peptides
appears reasonable, however there have only been limited
studies on the structural effect of the Arg side chain length
(Diss and Kennan 2008). Guanidinium-containing Arg
analogs with various chain lengths were incorporated at the
C-terminus of helical peptides (Schneider and DeGrado
1998). Optimal C-capping energetics was observed for
analogs with two and four methylenes (n = 1 and 3, Fig. 1)
(Schneider and DeGrado 1998). Interestingly, the molecule
with four methylenes can be viewed as an achiral version of
Arg, however analogs with even longer chain lengths were
not investigated. More complex effects of the Arg side chain
length on interhelical interactions were studied in a hetero-
dimeric coiled coil system (Diss and Kennan 2008). The Arg
analogs were incorporated at the helix–helix interface
a position (Diss and Kennan 2008). Although Arg could be
shortened by one methylene without affecting the stability
when incorporated into pairing monomers, shortening the
Arg side chain further resulted in significant decrease in
stability (DTm = –10ꢁC) (Diss and Kennan 2008). Also,
interhelical interactions between the Arg analogs and Asp/
Glu (both at the a position) were not significantly affected by
the side chain length (Diss and Kennan 2008). However, Arg
analogs with side chain lengths longer than Arg were not
investigated. These pioneering studies provided some initial
understanding on the structural effects of Arg analogs.
Importantly, many biological activities require specific
structures and conformations. Therefore, thorough studies
on the structural effect of incorporating Arg analogs would
serve as the foundation to facilitate the use of these analogs in
various applications. Since one-third of protein residues
adopt a helical conformation (Chou and Fasman 1974;
Barlow and Thornton 1988; Cheng et al. 2010), herein we
report the effect of Arg side chain length on helix formation
and capping. We incorporated Arg analogs with varying side
chain lengths into 19-residue peptides using Fmoc-based
chemistry. Furthermore, we investigated the helix formation
and capping parameters for the various Arg analogs by cir-
cular dichroism spectroscopy coupled with modified Lifson–
Roig theory, conformational analysis by molecular
mechanics calculations, and a survey of the protein structure
database.
One of the most direct methods to synthesize guanidi-
nium-containing Arg analogs would be to guanidinylate the
corresponding amine using one of the various guanidiny-
lation reagents (Robinson and Roskamp 1997; Shey and
Sun 1998; Ho and Sun 1999; Katritzky and Rogovoy
2005). Using these guanidinylation reagents, a number of
Arg analogs have been synthesized such as (S)-2-amino-6-
guanidinohexanoic acid (Agh; Fig. 1) (Poss et al. 1992;
Drake et al. 1994), a-(trifluoromethyl)- and a-(difluoro-
methyl)-arginines (Moroni et al. 2001), and cyclopropane-
containing Arg analogs (Pradhan et al. 2009). Furthermore,
solid phase guanidinylation has been used to prepare pep-
tides containing Agh (Fig. 1) (Yong et al. 1997, 1999;
Feichtinger et al. 1998a; Zhang and Kennan 2001), Agb
(Fig. 1) (Yong et al. 1997, 1999; Zhang and Kennan 2001;
Diss and Kennan 2008), and Agp (Fig. 1) (Diss and Ken-
nan 2008). Some of these Arg analog-containing peptides
have been prepared by 9-fluorenylmethoxycarbonyl
(Fmoc)-based chemistry, including Agb-containing dipep-
tide (Yong et al. 1997, 1999) and heptapeptide (Mitchell
et al. 2000), and Agh-containing tripeptide (Yong et al.
1997, 1999) and octapeptide (Feichtinger et al. 1998a).
However, peptides with more than ten residues have only
been prepared by Boc-based chemistry, including Agp- and
Agb-containing 30-mers (Zhang and Kennan 2001; Diss
O
O
H
N
NH
2
H N
3
n
NH
2
n = 1, Agp, (S)-2-amino-3-guanidinopropionic acid
2, Agb, (S)-2-amino-3-guanidinobutyric acid
3, Arg, L-arginine
4, Agh, (S)-2-amino-3-guanidinohexanoic acid
H
N
n
O
O
NH
2
N
H
Results
NH
n
3
H N
3
NH
2
n = 1, 2, 3
C-cap moieties
n = 4, Lys, L-lysine
Synthesis of protected Arg analogs
The appropriately protected Arg analogs suitable for Fmoc-
based solid phase peptide synthesis were synthesized
Fig. 1 Chemical structures of arginine, arginine analogs, lysine, and
aminoguanidinium C-cap moieties
123

Helix formation and capping energetics of arginine analogs
197
following known published procedures using N-triflylgua-
nidinine derivatives (Scheme 1) (Feichtinger et al. 1998a,
1998b). In short, the Na-Fmoc-protected Lys analog was
silated using methyl(trimethylsilyl)-trifluoroacetanimide in
refluxing dichloromethane to give a clear solution. Gua-
nidinylation was then performed by treatment with N,N0-di-
Boc-N00-triflylguanidine (Feichtinger et al. 1998a, 1998b)
at room temperature to give the desired products. The
resulting amino acid backbone amine was already pro-
tected with Fmoc and the side chain guanidine group was
protected with two Boc groups, suitable for Fmoc-based
solid phase peptide synthesis (Fields and Noble 1990).
surrounded by residues in random coil conformation). This
enables the exclusive derivation of the N-cap n parameter
without needing to consider the helix propensity w param-
eter for residue Xaa. Similarly, the CXaa19 peptides enable
the exclusive derivation of the C-cap c parameter without
needing to consider the helix propensity w parameter for
residue Xaa.
Peptides were synthesized by Fmoc-based chemistry
(Fields and Noble 1990). For the Arg residues in the Arg-
containing peptides, Fmoc–Arg(Pbf)–OH was double
coupled for 25 min, because intramolecular cyclization of
the activated ester occurs readily in this time frame,
effectively terminating the coupling reaction. For peptides
with Arg as the first residue coupled to the resin, quadruple
coupling was performed to ensure attachment to the resin.
For the Agh residues in the Agh-containing peptides,
Fmoc–Agh(Boc)2–OH was coupled for 45 min using
standard protocols. For peptides with Agh as the first res-
idue coupled to the resin, coupling was performed for 8 h
to ensure attachment to the resin. The lack of need for
double coupling is consistent with the unlikelihood of
intramolecular cyclization. Intramolecular cyclization of
the Agh activated ester would form a seven-membered
ring, a medium size ring, which is inherently more difficult
to form compared to six-membered rings.
Design and synthesis of Arg analog-containing
Ala-based peptides
The Ala-based peptides were designed according to anal-
ogous peptides initially studied by Baldwin (Chakrabartty
et al. 1994; Doig and Baldwin 1995) and later by our group
(Chiu et al. 2006; Cheng et al. 2010) (Table 1). To inves-
tigate the effect of the Arg side chain length on helix
propensity (w), four XaaAla peptides were designed with
four positions substituted simultaneously with the same
Arg analog (Chakrabartty et al. 1994). The four NXaa01
peptides and the four CXaa19 peptides (each with one Arg
analog at the N-terminus and C-terminus, respectively)
were designed to derive the N-cap (n) and C-cap
(c) parameters for the Arg analogs, respectively (Doig and
Baldwin 1995). The N-terminus of all peptides was acet-
ylated and the C-terminus was designed to be a carboxy-
amide so there would be no bias (from charged termini) on
the statistical mechanical capping parameters derived for
the Arg analogs (Cheng et al. 2010). Based on the modified
Lifson–Roig theory (Lifson and Roig 1961; Doig et al.
1994; Chakrabartty et al. 1994; Doig and Baldwin 1995;
Chiu et al. 2006; Cheng et al. 2010), the probability of
residue Xaa for all conformational states in NXaa01 pep-
tides would be represented by only v (initiation parameter),
n (N-cap parameter), and 1 (the probability for random coil
For the Agb residues in peptide AgbAla, Fmoc–
Agb(Boc)2–OH could not be used to introduce all four Agb
residues even with triple coupling or microwave condi-
tions. Instead, the corresponding Dab-containing peptide
was synthesized with Fmoc–Dab(ivDde)–OH (Dab,
(S)-2,4-diaminobutyric acid; ivDde, isovaleryldimedone;
Scheme 2) (Chhabra et al. 1998). The ivDde protecting
group was then selectively removed using hydrazine
followed by solid phase guanidinylation using N,N0-di-
Boc-N00-triflylguanidine in the presence of triethylamine
(Scheme 2). The corresponding Dab-containing peptide
synthesized using Fmoc–Dab(Mtt)–OH did not yield the
intended products; coupling of this residue appeared to be
difficult. For the Agp residues in peptide AgpAla, coupling
Scheme 1 Synthesis of
H
N
N
appropriately protected Arg
analogs for Fmoc-based solid
phase peptide synthesis
Boc
Boc
O
NH
NH
3
Si
O
O
F C
N
n
1.
, CH Cl
n
3
2
2
O
OH
O
N
H
O
N
H
NTf
O
O
2, Boc
Boc
N
N
H
H
n=1, Fmoc-Agp(Boc) -OH, 74%
2
2, Fmoc-Agb(Boc) -OH, 63%
2
4, Fmoc-Agh(Boc) -OH, 74%
2
123

198
R. P. Cheng et al.
Table 1 Sequence of Ala-
based peptides for determining
the N-cap parameter, C-cap
parameter, and helix propensity
of Arg analogs with varying
side chain length
Peptide
Sequence
NAgp01 Ac-Agp Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Gly Gly Tyr-NH2
NAgb01 Ac-Agb Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Gly Gly Tyr-NH2
NArg01 Ac-Arg Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Gly Gly Tyr-NH2
NAgh01 Ac-Agh Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Gly Gly Tyr-NH2
CAgp19 Ac-Tyr Gly Gly Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Agp-NH2
CAgb19 Ac-Tyr Gly Gly Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Agb-NH2
CArg19 Ac-Tyr Gly Gly Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Arg-NH2
CAgh19 Ac-Tyr Gly Gly Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Lys Ala Ala Ala Ala Agh-NH2
AgpAla Ac-Tyr Gly Gly Agp Ala Ala Ala Ala Agp Ala Ala Ala Ala Agp Ala Ala Ala Ala Agp-NH2
AgbAla Ac-Tyr Gly Gly Agb Ala Ala Ala Ala Agb Ala Ala Ala Ala Agb Ala Ala Ala Ala Agb-NH2
ArgAla
Ac-Tyr Gly Gly Arg Ala Ala Ala Ala Arg Ala Ala Ala Ala Arg Ala Ala Ala Ala Arg-NH2
The Arg analogs are shown in
bold
AghAla Ac-Tyr Gly Gly Agh Ala Ala Ala Ala Agh Ala Ala Ala Ala Agh Ala Ala Ala Ala Agh-NH2
using Fmoc–Agp(Boc)2–OH gave very small amounts of
desired peptide even with triple couple or microwave
conditions. This is most likely due to steric hindrance
caused by the close proximity of the Boc protecting groups
to the backbone because of the short side chain length.
Since insufficient material was obtained using Fmoc–
Agp(Boc)2–OH, a solid phase guanidinylation strategy was
adopted (Scheme 3). The corresponding Dap-containing
peptide was synthesized using Fmoc–Dap(Mtt)–OH (Dap,
(S)-2,3-diaminopropionic acid; Mtt, 4-methyltrityl), then
the Mtt protecting group was selectively removed using 1%
trifluoroacetic acid/methylene chloride followed by solid
phase guanidinylation using N,N0-di-Boc-N00-triflylguani-
dine in the presence of triethylamine under microwave
conditions. We did not explore the synthetic route
involving the ivDde protecting group for the Agp-con-
taining peptides, because side chain to main chain migra-
tion may occur with Dap(ivDde)-containing peptides
during solid phase synthesis.
O
O
NH
Protected Peptide with Dab(ivDde)
N
H
O
2% hydrazine/DMF
NH2
Protected Peptide with Dab
N
H
O
Tf
N
Boc
Boc
N
N
H
H
Upon cleavage with concomitant side chain deprotec-
tion, the peptides were confirmed by MALDI-TOF mass
spectrometry and purified by reverse phase high perfor-
mance liquid chromatography to [98% purity. The con-
centration of the peptides was determined by the Edelhoch
method (Edelhoch 1967; Pace et al. 1995). The peptides
bear multiple positive charges at neutral pH and the pep-
tides are not amphiphilic. As such, these peptides should
not associate in neutral aqueous buffers. Furthermore,
analogous peptides have been shown to be monomeric in
solution (Padmanabhan et al. 1990; Chakrabartty et al.
1994; Doig and Baldwin 1995; Chiu et al. 2006; Chiu and
Cheng 2007; Cheng et al. 2010), and the CD spectrum of
each peptide did not alter significantly between 80 and
160 lM. Therefore, the peptides are most likely mono-
meric in solution, and intermolecular interactions should
not contribute significantly to the helical content of these
peptides in solution.
Et3N, DCM, microwave
Boc
N
HN
NH
Boc
Protected Peptide with Agb
N
H
O
Scheme 2 Synthesis of Agb-containing peptides by solid phase
guanidinylation
Circular dichroism spectroscopy of the Arg
analog-containing peptides
The circular dichroism (CD) spectrum for all peptides was
acquired at pH 7 in the presence of 1 M NaCl to derive the
fraction helix (fhelix) (Table 2 and Fig. 2); these were the
123

Helix formation and capping energetics of arginine analogs
199
(Table 3). The corresponding free energies were also
derived (Table 3). The N-cap n parameter was less than
zero for all Arg analogs, suggesting extremely unfavorable
energetics for placing these positively charged residues at
the N-terminus. However, values less than zero do not bear
physical meaning, therefore the n values were set to zero.
The C-cap c parameter followed the trend Agh [ Arg [
Agb [ Agp (Table 3). In other words, the c values
increased with increasing side chain length. Interestingly,
only Agh was clearly energetically favorable at the C-cap
position, whereas Arg was indifferent; Agb was somewhat
energetically unfavorable, but Agp was very unfavorable.
The helix propensity w followed the trend Agp \ Agb \
Arg [ Agh. Only the longer Arg and Agh were energeti-
cally favorable in the helical conformation, whereas the
shorter Agb and Agp were unfavorable. However, Arg was
more energetically favored in the helical conformation than
Agh, highlighting the uniqueness of the Arg side chain
length in helix formation.
HN
Protected Peptide with Dap(Mtt)
N
H
O
1% TFA/DCM
H2N
Protected Peptide with Dap
N
H
O
Tf
N
Boc
Boc
N
N
H
H
Et3N, DCM, microwave
Conformational analysis by molecular mechanics
calculations
H
N
N
Boc
Boc
O
HN
Conformational analysis was performed by molecular
mechanics calculations to gain further insight into the helix
capping and formation energetic trends of the Arg analogs
with varying side chain length (Table 4). The peptides
were composed of 20 alanines and one Arg analog at
various positions. Each Arg analog was incorporated at the
first four positions (N1–N4), the last four positions (C4–
C1), and the center position of a 21-residue helical peptide.
For all peptide models, the backbone was fixed to ideal
helix dihedrals (/ = -57ꢁ, w = -47ꢁ) (Edsall et al. 1966)
to avoid drastic changes in the backbone conformation
especially at the termini of the helix. For a given peptide,
different side chain conformations were generated using
various v angle combinations. For each v angle involving
only sp3 atoms such as v1, three possible low-energy
staggered conformations were considered (Fig. 3):
gauche- (60ꢁ, g-), trans (180ꢁ, t), and gauche? (300ꢁ, g?)
(McGregor et al. 1987; Dunbrack and Karplus 1993). For
v angles involving the planar guanidinium nitrogen (due to
sp2 hybridization of the nitrogen and carbon atoms), 12
different angles were considered: 0ꢁ, 30ꢁ, 60ꢁ, 90ꢁ, 120ꢁ,
150ꢁ, 180ꢁ, 210ꢁ, 240ꢁ, 270ꢁ, 300ꢁ, and 330ꢁ. A combined
total of 12,960 conformations were minimized using the
CFF forcefield and analyzed.
Protected Peptide with Agp
N
H
Scheme 3 Synthesis of Agp-containing peptides by solid phase
guanidinylation
conditions for the experiments performed by Baldwin and
coworkers (Chakrabartty et al. 1994). The mean residue
ellipticity at 222 nm reflects the helical content of a given
peptide. More specifically, the more negative the CD signal
at 222 nm, the more helical the peptide. The helical content
of the NXaa01 peptides followed the trend NAgh01*
NArg01*NAgb01 [ NAgp01 (Fig. 2a). This trend sug-
gests similar N-capping abilities for the Arg analogs except
Agp, which showed slightly weaker N-cap capabilities. The
helical content of the CXaa19 peptides were quite similar
but followed the trend CAgh19 [ CArg19 [ CAgb19 [
CAgp19 (Fig. 2b). This trend suggests that C-capping
ability for the Arg analogs increases with increasing side
chain length. The helical content of the XaaAla peptides
followed the trend ArgAla [ AghAla [ AgbAla [ Ag-
pAla (Fig. 2c).
Helix formation parameters for Arg analogs
The energy of the lowest energy conformations for
peptides with the same Arg analog at different positions
was then compared to gain information regarding the
positional preference (Table 4). Incorporating a given Arg
analog near the helix N-terminus resulted in higher energy
conformers compared to incorporating the same Arg
The helix formation parameters including N-cap (n), C-cap
(c), and helix propensity (w) were derived from the CD
data using modified Lifson–Roig theory (Lifson and Roig
1961; Doig et al. 1994; Chakrabartty et al. 1994; Doig and
Baldwin 1995; Chiu et al. 2006; Cheng et al. 2010)
123

200
R. P. Cheng et al.
analog near the C-terminus or at the center of the helix.
Furthermore, the energy difference between positioning a
given Arg analog at N1 and C1/center was smaller for the
shorter residues Agp and Agb compared to the longer
residues Arg and Agh. These position-dependent energy
trends are consistent with the experimental results (vide
supra), showing unfavorable energetics for the N-cap
parameter (n) of the Arg analogs compared to the C-cap
parameter (c) and helix propensity (w).
Table 2 Mean residue ellipticity at 222 nm and fraction helix (fhelix
of Ala-based peptides containing Arg analogs with varying side chain
length
)
Peptide
[h]222
fhelix
NAgp01
NAgb01
NArg01
NAgh01
CAgp19
CAgb19
CArg19
CAgh19
AgpAla
AgbAla
ArgAla
AghAla
-15800 200
-18200 200
-18500 200
-18000 200
-17000 200
-17500 200
-18100 200
-18400 200
-5600 200
-16000 200
-24400 200
-21000 200
0.452 0.006
0.520 0.006
0.529 0.006
0.514 0.006
0.485 0.004
0.500 0.005
0.517 0.005
0.529 0.004
0.161 0.007
0.463 0.005
0.697 0.006
0.594 0.006
Conformations within 4 kcal of the lowest energy con-
former for each peptide were then examined in more detail
(Table 4), because room temperature can provide up to
4 kcal mol-1 of thermal energy. The positional depen-
dence for the weighed average energy (based on Boltz-
mann distribution) of the low energy ensembles followed
the same trend as the energy for the lowest energy con-
formations. Interestingly, the number of conformations at
positions N1 and N2 was consistently higher than the
number of conformations at the other positions for peptides
containing Agb, Arg, and Agh. This positional dependence
in low energy conformers was barely present for Agp-
containing peptides, perhaps because there were not many
low energy conformers thereby suppressing the difference
in number of conformations. Disregarding the Agp pep-
tides, there were no low energy structures with g– con-
formation at the v1 dihedral except for the N1 and N2
positions. This lack of the g– conformation (at positions
other than N1 and N2) is most likely because the Ca–Cb
vector inherently points toward the N-terminus of a helix,
and the g– conformation orients the side chain towards the
upstream backbone resulting in unfavorable steric interac-
tions with the i-3 position (Fig. 3). This unfavorable steric
interaction remains present at the N3 and N4 positions, but
is alleviated at the N1 and N2 positions due to the absence
of the upstream i-3 residue. As for Agp, the side chain is so
short that the guanidinium group is directly attached to the
Cb carbon. The planar geometry of the guanidinium group
inherently directs the side chain away from the backbone,
thereby avoiding this unfavorable steric clash. As such, the
g– conformation at the v1 dihedral is present in all the low
energy conformers of the Agp-containing peptides. To
further confirm these molecular mechanics results, we
performed a survey on natural protein structures.
Survey of protein structures
The non-redundant protein structure database PDBselect
(April 2009, 25% threshold) (Hobohm and Sander 1994;
Griep and Hobohm 2010) was surveyed for Arg in the
context of various helix-related structures (Table 5). The
definition of the helical conformation was based on back-
bone dihedrals (Gunasekaran et al. 1998; Engel and
DeGrado 2004; Cheng et al. 2010), and only helices with at
Fig. 2 Circular dichroism spectra of the peptides at pH 7 (273 K) in
1 mM phosphate, borate, and citrate buffer with 1 M NaCl.
a NAgp01, NAgb01, NArg01, NAgh01. b CAgp19, CAgb19,
CArg19, CAgh19. c AgpAla, AgbAla, ArgAla, AghAla
123

Helix formation and capping energetics of arginine analogs
201
Table 3 Statistical mechanical helix formation parameters for Arg analogs with varying side chain length derived from experimentally mea-
sured fraction helix based on modified Lifson–Roig theory
Residue
n
c
w
DGbC-cap
DGchelix
Agp
Agb
Arg
Agh
0a
0a
0a
0a
0.0730 0.1036
0.497 0.155
1.06 0.18
0.289 0.035
0.845 0.039
1.68 0.08
1.42
0.674
0.0914
0.380
-0.0316
-0.213
-0.281
-0.0944
1.48 0.16
1.19 0.06
These statistical mechanical helix formation parameters are derived from experimental CD data of the peptides in Table 2; for sequences of the
peptides see Table 1
a
The parameter initially converged to a negative probability (which carries no physical meaning), therefore the value was set to zero
DGC-cap = -RT ln (c) reported in kcal mol-1
DGhelix = -RT ln (w) reported in kcal mol-1
b
c
least six residues were considered to be a helix to avoid
‘‘helices’’ less than one turn (Cheng et al. 2007, 2010). A
total of 4,418 protein chains and 666,086 residues were
considered involving 17,622 helices and 236,790 helical
residues. Approximately 33.7% of the residues in the
database were helical (236,790 out of 666,086) (Cheng et al.
2010), similar to earlier surveys (Chou and Fasman 1974;
Barlow and Thornton 1988). Interestingly, 44.4% of the Arg
residues in the database were helical (14,866 out of 33,447)
with a statistical propensity of 1.20 0.01 (Z value 22.01;
P value 2.31 9 10-107). This result is different from an
earlier survey (Chou and Fasman 1974), which found the
statistical helix propensity for Arg to be 0.79, but similar to
a more recent survey (Kumar and Bansal 1998), which
found the statistical propensity for Arg to be 1.33. To gauge
the significance of the occurrences for Arg at various
positions in a helix, the statistical propensity was derived by
dividing the occurrence by the corresponding expected
occurrence. The expected occurrence was obtained by
bootstrapping (Efron and Gong 1983) the entire database
including all structures. Bootstrapping was performed
100,000 times for these cases to obtain standard deviations
for the expected occurrence. This enabled the calculation of
standard deviations for the statistical propensities and
Z values (Table 5), which was used to derive the P values
(Klugh 1970; Kuebler and Smith 1976). Statistical pro-
pensities greater than unity represent higher occurrence
than expected based on occurrence in all structures, whereas
propensities less than unity indicate lower occurrence than
the expected value. To survey Arg at various positions in a
helix, we considered positions near the N-terminus, near the
C-terminus, and internal positions. The residues near the
helix termini were designated as follows:
This definition is similar but not identical to previous
definitions for surveying helix capping (Presta and Rose
1988; Richardson and Richardson 1988; Aurora and Rose
1998; Gunasekaran et al. 1998).
The propensity for Arg to appear as the first three resi-
dues in a helix was clearly less than unity (Table 5), con-
sistent with potentially repulsive interactions between the
positively charged Arg side chain and the partial positive
charge of the helix macrodipole (Blagdon and Goodman
1975; Sali et al. 1988). Furthermore, the propensity for the
three positions prior to the helix was also less than unity,
most likely due to unfavorable interactions with the helix
macrodipole even though the positions are no longer in the
helix, or perhaps inherently lower incorporation of Arg in
non-helical structures. The propensity for Arg in the last
three positions of a helix was higher than unity, consistent
with potentially favorable interactions between the posi-
tively charged Arg side chain and the partial negative
charge of the helix macrodipole (Blagdon and Goodman
1975; Sali et al. 1988). In contrast, the propensity for the
three positions after the helix was less than unity, most
likely due to the inherently lower incorporation of Arg in
non-helical structures.
The side chain conformation of the Arg residues was
then examined in detail (Table 6). Residues at positions
upstream (N, N0, N00) and downstream (C, C0, C00) of the
helix are connected to the helix by residues with non-
helical backbone dihedrals at positions N and C, respec-
tively. As such, Arg at these positions beyond the helix was
not included in this analysis. Since the total occurrence of
Arg at the different positions is different, the percentage for
the three low energy conformations of the v1 dihedral at
each position was derived to enable more meaningful
comparisons (Table 6). For the v1 dihedral, the g– con-
formation occurred less than t or g? at all positions, con-
sistent with our conformational analysis by molecular
mechanics calculations. This distribution in conformation
is most likely due to sterics (vide supra; Fig. 3). However,
the non-ideal helical dihedrals in natural proteins may
N00–N0–N–N1–N2–N3–…–C3–C2–C1–C–C0–C00
The first residue of a helix with helical dihedral angles is
N1, and the immediate preceding residue with non-helical
dihedrals is designated N. The last residue of a helix with
helical dihedral angles is C1, and the immediate down-
stream residue with non-helical dihedrals is designated C.
123

202
R. P. Cheng et al.
Table 4 Low energy
conformations from
conformational analysis of Arg
analog-containing peptides by
molecular mechanics
calculations
Residue
Position
Lowest energy
conformer
Energy (kcal)
Conformations within 4 kcal of lowest
Number of
conformationsa
v1b
g-
t
g?
Agp
Agp
Agp
Agp
Agp
Agp
Agp
Agp
Agp
Agb
Agb
Agb
Agb
Agb
Agb
Agb
Agb
Agb
Arg
Arg
Arg
Arg
Arg
Arg
Arg
Arg
Arg
Agh
Agh
Agh
Agh
Agh
Agh
Agh
Agh
Agh
N1
-33.7
-33.8
-34.3
-34.7
-35.3
-35.5
-35.6
-35.6
-36.0
-35.0
-35.2
-35.7
-36.3
-36.5
-36.7
-36.7
-36.8
-37.4
-37.9
-38.1
-39.3
-39.7
-40.2
-40.4
-40.4
-40.4
-40.8
-41.0
-41.3
-42.4
-43.0
-43.7
-43.9
-43.8
-43.9
-44.1
15
14
4
4
5
4
4
5
3
4
4
4
5
6
6
N2
N3
12
3
5
N4
13
3
5
Center
C4
11
3
5
12
3
5
C3
12
3
5
C2
12
3
5
C1
15
4
6
N1
44
14
14
0
16
16
14
16
17
16
15
15
14
43
42
29
30
26
27
25
27
18
103
97
74
70
47
42
37
41
31
14
14
16
16
17
17
16
18
18
44
47
40
30
29
31
29
31
30
112
118
104
85
77
74
71
72
82
N2
44
N3
30
N4
32
0
Center
C4
34
0
33
0
C3
31
0
C2
33
0
C1
32
0
N1
113
118
69
26
29
0
N2
N3
N4
60
0
Center
C4
55
0
58
0
C3
54
0
C2
58
0
C1
48
0
N1
281
292
178
155
124
116
108
113
113
66
77
0
N2
a
The number of conformations
N3
within 4 kcal of the lowest
energy conformer for each
peptide
N4
0
Center
C4
0
b
0
The number of conformations
for the three staggered v1
C3
0
dihedrals (Fig. 3): g- (gauche–,
v1 = 60ꢁ), t (trans, v1 = 180ꢁ),
g? (gauche?, v1 = 300ꢁ)
C2
0
C1
0
avoid this unfavorable steric interaction, leading to some
representation in the survey. Low occurrence of g– was
particularly significant at the internal, C3, and N3 posi-
tions, with slightly higher percentage of g– for the positions
near the helix termini (C1, C2, N1, N2), most likely due to
end fraying effects. Importantly, there is a higher per-
centage of g– for the positions N1 and N2 compared to C1
and C2. This can be attributed to the absence of the i-3
helical residue for residues at positions N1 and N2,
inherently avoiding unfavorable steric interactions (Fig. 3).
Discussion
All Arg analogs, regardless of side chain length, exhibited
extremely unfavorable energetics at the N-cap position of
a helix (i.e. n parameter; Table 3). This may be due to
destabilizing interactions with the partial positive charge
of the helix macrodipole at the N-terminal end (Blagdon
and Goodman 1975; Sali et al. 1988). Accordingly, one
would a priori predict that positively charged amino acids
should be favored at the C-cap position (i.e. c parameter)
123

Helix formation and capping energetics of arginine analogs
203
due to stabilizing interactions with the partial negative
charge of the helix macrodipole (Blagdon and Goodman
1975; Sali et al. 1988). However, only Agh was ener-
getically favorable at the C-cap position. Arg was ener-
getically indifferent, but Agb and Agp were energetically
unfavorable. Apparently, short analogs of Arg cannot
provide stabilization through interaction with the helix
macrodipole at the C-terminal end of a helix, perhaps due
to the need to adopt alternative non-helical conformations
for the favorable interaction. Regardless, the c parameter
followed the trend Agh [ Arg [ Agb [ Agp (Table 3).
This trend is not exactly the same as the findings by
(Schneider and DeGrado 1998) on linear alkyl amino-
guanidines, most likely because of the lack of the car-
boxyamide group and stereocenter in the DeGrado study.
Interestingly, the helix propensity w followed the trend
Agp \ Agb \ Arg [ Agh (Table 3). This is in sharp
contrast with previous studies on positively charged Lys
analogs (Padmanabhan et al. 1996), showing that helix
propensity w increases with increasing side chain length
(Padmanabhan et al. 1996). Similarly, the helix propensity
of linear hydrophobic residues also increases with
increasing side chain length (Padmanabhan and Baldwin
1991). Accordingly, our result on helix propensity high-
lights the uniqueness of Arg side chain length in favoring
helix formation.
Conformational analysis by molecular mechanics calcu-
lations showed that incorporating a given Arg analog at the
N-terminus of a helix is higher in energy (and thus less sta-
ble) compared to incorporating at the center or C-terminus
(Table 4), consistent with our experimentally derived
N-capping, C-capping, and helix propensity energetics
(Table 3). Furthermore, the difference between the energy of
Fig. 3 a Newman projections of the three low energy v1 dihedrals
gauche–, trans, and gauche? b Three-dimensional model of a generic
helix backbone with the three low energy v1 dihedrals highlighted for
the center residue
Table 5 Survey of arginine in helices and near the termini of helices with at least six residues in protein structures
Position
Occurrencea
Propensityb
Z valuec
P valued
N00
N0
736
660
0.884 0.030
0.780 0.026
0.548 0.018
0.946 0.031
0.920 0.030
0.789 0.026
1.247 0.015
1.41 0.05
-3.45
-6.57
-13.7
-1.64
-2.47
-6.47
20.53
12.5
5.82 9 10-4
5.03 9 10-11
8.85 9 10-43
1.01 9 10-1
1.35 9 10-2
9.80 9 10-11
1.16 9 10-93
5.80 9 10-36
8.54 9 10-51
1.59 9 10-14
3.02 9 10-8
1.91 9 10-4
1.50 9 10-1
N
485
N1
N2
N3
Internal
C3
C2
C1
C
835
813
696
8,167
1,243
1,318
1,103
723
1.49 0.05
15.0
1.25 0.04
7.68
0.819 0.027
0.874 0.029
0.951 0.032
-5.54
-3.73
-1.44
C0
C00
723
768
a
The occurrence of Arg at the designated position (in helices and near the termini of helices) in the non-redundant protein structure database
PDBselect (Hobohm and Sander 1994; Griep and Hobohm 2010) (April 2009, 25% threshold). Only helices with at least six consecutive helical
residues were considered to avoid counting any given residue more than once due to proximity to both N- and C-termini
b
The occurrence divided by the corresponding expected value. The expected value was obtained by bootstrapping the entire PDBselect structure
database 100,000 times
c
The difference between the occurrence and the expected value divided by the SD for the expected value
d
The probability that the occurrence and expected occurrence are the same based on the SD obtained from bootstrapping assuming a Gaussian
distribution
123

204
R. P. Cheng et al.
the stability of non-helix structures such as sheets and turns
would be necessary for a more complete understanding of
the structural effect of the Arg side chain length, to enable
facile use of Arg analogs in various applications.
Table 6 Occurrence and percent occurrence of v1 dihedral confor-
mations of arginine in helices and near the termini of helices with at
least six residues in protein structures
Position
v1
g–
t
g?
308 (38%)
N1
88 (11%)
409 (51%)
334 (43%)
336 (49%)
Conclusions
N2
87 (11%)
23 (3%)
358 (46%)
321 (47%)
4079 (51%)
667 (55%)
713 (55%)
850 (78%)
N3
We have demonstrated the synthesis of 19-residue peptides
containing Arg analogs with varying side chain length at
different positions using Fmoc-based chemistry. Straight-
forward solid phase peptide synthesis was sufficient for
incorporating Agh, whereas solid phase guanidinylation
using orthogonal protection strategies were necessary for
Agb and Agp-containing peptides. Importantly, the effect
of Arg side chain length on helix formation and capping
was investigated in Ala-based peptides by CD. All four Arg
analogs were energetically unfavorable for N-capping.
Interestingly, Arg was the most energetically favorable at
internal helix positions, whereas Agh was the most ener-
getically favorable for C-capping. Agb and Agp were not
suitable for helical structures, but may be useful for sheets
or turns. Molecular mechanics calculations and a survey on
a non-redundant protein structure database were consistent
with our experimental findings, and provided conforma-
tional insight into the experimentally observed trends. The
calculations and survey both showed that the g– confor-
mation for the v1 dihedral is present for the first two resi-
dues at the N-terminus of helices, but not favored in the
center or the C-terminus of helices. These thorough studies
on the effect of Arg side chain length on helix formation
and capping should serve as the foundation for the use of
Arg analogs in developing Arg-related bioactive com-
pounds and technologies.
Internal
C3
247 (3%)
56 (5%)
3625 (46%)
496 (41%)
473 (37%)
147 (14%)
C2
107 (8%)
90 (8%)
C1
The occurrence of Arg at the designated position (in helices and near
the termini of helices) in the non-redundant protein structure database
PDBselect (Hobohm and Sander 1994; Griep and Hobohm 2010)
(April 2009, 25% threshold). Only helices with at least six consecu-
tive helical residues were considered to avoid counting any given
residue more than once due to proximity to both N- and C-termini.
The percent occurrence is presented in parentheses. The percent
occurrence for each conformation at a given position was calculated
by dividing the occurrence by the sum of the occurrences for all three
conformations for that given position
incorporating a given Arg analog at the N-terminus and
C-terminus (or center) increases with increasing side chain
length (Table 4), again consistent with our experimental
results (Table 3). However, the slight energetic difference
between the experimentally determined helix propensity and
C-cap energetics (less than 1 kcal mol-1) could not be
clearly distinguished with the conformational analysis using
molecular mechanics calculations as performed in this study.
Although this could be due to fixing the backbone to an ideal
helix conformation, results from the same calculation with-
out fixing the backbone gave similar energetic differences.
Another possibility is that the simple conformational anal-
ysis was performed using an implicit solvent model. More
accurate modeling involving explicit water may be neces-
sary to capture the subtle difference between helix propen-
sity and C-cap parameter. However, the inherent error for
these calculations may be too large to distinguish the ener-
getic difference between the helix propensity and C-cap
parameter. Regardless, the g– conformation for the v1
dihedral is not present except at the N1 or N2 position, or in
the Agp-containing peptides. Importantly, these conforma-
tional results from molecular mechanics calculations are
consistent with the survey results on natural protein
structures.
Acknowledgments This work was supported by NYSTAR James
D. Watson Investigator Program (R.P.C.), Kapoor funds (R.P.C.), The
State University of New York at Buffalo (R.P.C.), National Science
Foundation (R.P.C., CHE0809633), National Taiwan University
(R.P.C.), National Science Council (R.P.C., NSC-97-2113-M-002-
019-MY2, NSC-98-2119-M-002-025, NSC-99-2113-M-002-002-
MY2; Y.J.W., NSC-98-2815-C-002-056-M). The authors would like
to thank the Computer and Information Networking Center at
National Taiwan University for the support of the high-performance
computing facilities. The authors would like to thank Professor
Cheu-Pyeng Cheng (Department of Chemistry, National Tsing Hua
University) for helpful discussions, and Ms. Chia-Wen Kuo for
proofreading the manuscript.
We have focused on helix formation energetics includ-
ing N-capping, C-capping and helix propensity in this
study. However, protein secondary structures include
helices, sheets, and turns. Although 44.4% of the Arg
residues adopt a helical conformation (vide supra), the
other structures cannot be ignored. Therefore, studies on
References
Argos P (1988) An investigation of protein subunit and domain
interfaces. Protein Eng 2:101–113
Aurora R, Rose GD (1998) Helix capping. Protein Sci 7:21–38
Barlow DJ, Thornton JM (1988) Helix geometry in proteins. J Mol
Biol 201(3):601–619
123

Helix formation and capping energetics of arginine analogs
205
Baumgart S, Lindner Y, Ku¨hne R, Oberemm A, Wenschuh H, Krause
E (2004) The contributions of specific amino acid side chains to
signal intensities of peptides in matrix-assisted laser desorption/
ionization mass spectrometry. Rapid Commun Mass Spectrom
18:863–868
Blagdon DE, Goodman M (1975) Mechanisms of protein and
polypeptide helix initiation. Biopolymers 14:241–245
Blaum BS, Deakin JA, Johansson CM, Herbert AP, Barlow PN, Lyon
Feichtinger K, Sings HL, Baker TJ, Matthews K, Goodman M
(1998a) Triurethane-protected guanidines and triflyldiurethane-
protected guanidines: new reagents for guanidinylation reac-
tions. J Org Chem 63:8432–8439
Feichtinger K, Zapf C, Sings HL, Goodman M (1998b) Diprotected
triflyguanidines: a new class of guanidinylation reagents. J Org
Chem 63:3804–3805
Fields GB, Noble RL (1990) Solid-phase peptide-synthesis utilizing
9-fluorenylmethoxycarbonyl amino-acids. Int J Pept Protein Res
35(3):161–214
Griep S, Hobohm U (2010) PDBselect 1992–2009 and PDBfilter-
select. Nucleic Acids Res 38:D318–D319
Gunasekaran K, Nagarajaram HA, Ramakrishnan C, Balaram P
(1998) Stereochemical punctuation marks in protein structures:
glycine and proline containing helix stop signals. J Mol Biol
275:917–932
Hirsch AKH, Fischer FR, Diederich F (2007) Phosphate recognition
in structural biology. Angew Chem Int Ed 46:338–352
Ho K-C, Sun C-M (1999) Liquid phase parallel synthesis of
guanidines. Bioorg Med Chem Lett 9:1517–1520
M, Uhrin
D (2010) Lysine and arginine side chains in
glycosaminoglycan–protein complexes investigated by NMR,
cross-linking, and mass spectrometry: a case study of the factor
H-heparin interaction. J Am Chem Soc 132:6374–6381
Bogan AA, Thorn KS (1998) Anatomy of hot spots in protein
interfaces. J Mol Biol 280:1–9
Chakrabarti P (1994) Conformations of arginine and lysine side chains
in association with anions. Int J Peptide Protein Res 43:284–291
Chakrabartty A, Kortemme T, Baldwin RL (1994) Helix propensities of
the amino-acids measured inalanine-based peptides without helix-
stabilizing side-chain interactions. Protein Sci 3(5):843–852
Cheng RP, Girinath P, Ahmad R (2007) Effect of lysine side chain
length on intra-helical glutamate–lysine ion pairing interactions.
Biochemistry 46:10528–10537
Hobohm U, Sander C (1994) Enlarged representative set of protein
structures. Protein Sci 3(3):522–524
Cheng RP, Girinath P, Suzuki Y, Kuo H-T, Hsu H-C, Wang W-R,
Yang P-A, Gullickson D, Wu C-H, Koyack MJ, Chiu H-P, Weng
Y-J, Hart P, Kokona B, Fairman R, Lin T-E, Barrett O (2010)
Positional effects on helical Ala-based peptides. Biochemistry
49:9372–9384
Chhabra SR, Hothi B, Evans DJ, White PD, Bycroft BW, Chan WC
(1998) An appraisal of new variants of Dde amine protecting
group for solid phase peptide synthesis. Tetrahedron Lett
39:1603–1606
Hu Z, Ma B, Wolfson H, Nussinov R (2000) Conservation of polar
residues as hot spots at protein interfaces. Proteins Struct Funct
Genet 39:331–342
Janin J, Miller S, Chothia C (1988) Surface, subunit interfaces and
interior of oligomeric proteins. J Mol Biol 204:155–164
Jones S, Thornton JM (1995) Protein–protein interactions: a review of
protein dimer structures. Prog Biophys Mol Biol 63:31–65
Katritzky AR, Rogovoy BV (2005) Recent developments in guany-
lating agents. ARKIVOC iv:49–87
Chiu H-P, Cheng RP (2007) Chemoenzymatic synthesis of (S)-
hexafluoroleucine and (S)-tetrafluoroleucine. Org Lett 9:5517–
5520
Klugh HE (1970) Statistics-the essentials for research. In: Statistics-
The essentials for research. John Wiley & Sons Inc, New York.
pp 350
Chiu HP, Suzuki Y, Gullickson D, Ahmad R, Kokona B, Fairman R,
Cheng RP (2006) Helix propensity of highly fluorinated amino
acids. J Am Chem Soc 128(49):15556–15557
Chou PY, Fasman GD (1974) Conformational parameters for amino
acids in helical, b-sheet, and random coil regions calculated from
proteins. Biochemistry 13(2):211–222
Diss ML, Kennan AJ (2008) Orthogonal recognition in dimeric coiled
coils via buried polar-group modulation. J Am Chem Soc
130:1321–1327
Doig AJ, Baldwin RL (1995) N- and C-capping preferences for all 20
amino acids in a-helical peptides. Protein Sci 4:1325–1336
Doig AJ, Chakrabartty A, Klingler TM, Baldwin RL (1994)
Determination of free-energies of N-capping in a-helices by
modification of the Lifson–Roig helix-coil theory to include
N-capping and C-capping. Biochemistry 33(11):3396–3403
Drake B, Patek M, Lebl M (1994) A convenient preparation of
monosubstituted N,N’-di(Boc)-protected guanidines. Synthesis:
579–582
Dunbrack RL, Karplus M (1993) Backbone-dependent rotamer library
for proteins—application to side-chain prediction. J Mol Biol
230(2):543–574
Edelhoch H (1967) Spectroscopic determination of tryptophan and
tyrosine in proteins. Biochemistry 6(7):1948–1954
Kuebler RR, Smith H (1976) Statistics-A beginning. In: Statistics-A
beginning. John Wiley & Sons, Inc, New York. pp 302
Kumar S, Bansal M (1998) Dissecting a-helices: position-specific
analysis of a-helices in globular proteins. Proteins Struct Funct
Genet 31:460–476
Lifson S, Roig A (1961) The theory of helix-coil transition in
polypeptides. J Chem Phys 34:1963–1974
McGregor MJ, Islam SA, Sternberg MJE (1987) Analysis of the
relationship between side-chain conformation and secondary
structure in globular-proteins. J Mol Biol 198(2):295–310
Mitchell DJ, Kim DT, Steinman L, Fathman CG, Rothbard JB (2000)
Polyarginine enters cells more efficiently than other polycationic
homopolymers. J Pept Res 56:318–325
Moroni M, Koksch B, Osipov SN, Crucianelli M, Frigerio M, Bravo
P, Burger K (2001) First synthesis of totally orthogonal protected
a-(trifluoromethyl)- and a-(difluoromethyl)arginines.
Chem 66:130–133
J Org
Nishikaze T, Takayama M (2006) Cooperative effect of factors
governing molecular ion yields in desorption/ionization mass
spectrometry. Rapid Commun Mass Spectrom 20:376–382
Northern TR, Lee J-C, Hoang L, Raymond J, Hwang D-R, Yannone
SM, Wong CH, Siuzdak G (2008) A nanostructure-initiator mass
spectrometry-based enzyme activity assay. Proc Natl Acad Sci
USA 105:3678–3683
´
Edsall JT, Flory PJ, Kendrew JC, Liquori AM, Nemethy G,
Ramachandran GN, Scheraga HA (1966) A proposal of standard
conventions and nomenclature for the description of polypeptide
conformations. J Mol Biol 15:399–407
Pace CN, Vajdos F, Fee L, Grimsley G, Gray T (1995) How to
measure and predict the molar absorption-coefficient of a
protein. Protein Sci 4(11):2411–2423
Efron B, Gong G (1983) A leisurely look at the bootstrap, the
jackknife, and cross-validation. Am Stat 37(1):36–48
Engel DE, De Grado WF (2004) Amino Acid propensities are
position-dependent throughout the length of a-helices. J Mol
Biol 337:1195–1205
Padmanabhan S, Baldwin RL (1991) Straight-chain non-polar amino
acids are good helix-formers in water. J Mol Biol 219:135–137
Padmanabhan S, Marqusee S, Ridgeway T, Laue TM, Baldwin Robert
L (1990) Relative helix-forming tendencies of nonpolar amino
acids. Nature 344:268–270
123

206
R. P. Cheng et al.
Padmanabhan S, York EJ, Stewart JM, Baldwin RL (1996) Helix
propensities of basic amino acids increase with the length of the
side-chain. J Mol Biol 257(3):726–734
Tsai C-J, Lin SL, Wolfson HJ, Nussinov R (1997) Studies of protein–
protein interfaces: a statistical analysis of the hydrophobic effect.
Protein Sci 6:53–64
Poss MA, Iwanowicz E, Reid JA, Lin J, Gu Z (1992) A mild and
efficient method for the preparation of guanidines. Tetrahedron
Lett 40:5933–5936
Umezawa N, Gelman MA, Haigis MC, Raines RT, Gellman SH
(2001) Translocation of a b-peptide across cell membranes.
J Am Chem Soc 124:368–369
Pradhan TK, Joosten A, Vasse J-L, Bertus P, Karoyan P, Szymoniak J
(2009) A concise stereoselective synthesis of 2-substituted
Wender PA, Mitchell DJ, Pattabiraman K, Pelkey ET, Steinman L,
Rothbard JB (2000) The design, synthesis, and evaluation of
molecules that enable or enhance cellular uptake: peptoid
molecular transporters. Proc Natl Acad Sci USA 97: 13003–13008
Wender PA, Rothbard JB, Jessop TC, Kreider EL, Wylie BL (2002)
Oligocarbamate molecular transporters: design, synthesis, and
biological evaluation of a new class of transporters for drug
delivery. J Am Chem Soc 124:13382–13383
1-aminocyclopropanecarboxyllic acids. Eur
5072–5078
J Org Chem
Presta LG, Rose GD (1988) Helix signals in proteins. Science
240(4859):1632–1641
Richardson JS, Richardson DC (1988) Amino acid preferences for
specific locations at the ends of a helices. Science 240:1648–
1652
Wender PA, Galliber WC, Goun EA, Jones JR, Pillow TH (2008) The
design of guanidinium-rich transporters and their internalization
mechanisms. Adv Drug Deliv Rev 60:452–472
Robinson S, Roskamp EJ (1997) Solid phase synthesis of guanidines.
Tetrahedron 53:6697–6705
Sali D, Bycroft M, Fersht AR (1988) Stabilization of protein structure
by interaction of alpha-helix dipole with a charged side chain.
Nature 355:740–743
Yong YF, Kowalski JA, Lipton MA (1997) Facile and efficient
guanylation of amines using thioureas and Mukaiyama’s reagent.
J Org Chem 62:1540–1542
Schneider JP, DeGrado WF (1998) The design of efficient a-helical
C-capping auxiliaries. J Am Chem Soc 120:2764–2767
Shey J-Y, Sun C-M (1998) Soluble polymer-supported synthesis of
N,N-di(Boc)-protected guanidines. Synlett 1423–1425
Shinohara Y, Furukawa J, Niikura K, Miura N, Nishimura S-I (2004)
Direct N-glycan profiling in the presence of tryptic peptides on
MALDI-TOF by controlled ion enhancement and suppression
upon glycan-selective derivatization. Anal Chem 76:6989–6997
Yong YF, Kowalski JA, Thoen JC, Lipton MA (1999) A new reagent
for solid and solution phase synthesis of protected guanidines
from amines. Tetrahedron Lett 40:53–56
Zhang Y, Kennan AJ (2001) Efficient introduction of protected
guanidines in BOC solid phase peptide synthesis. Org Lett
3:2341–2344
Zhao Y, Kent SBH, Chait BT (1997) Rapid, sensitive structure analysis
of oligosaccharides. Proc Natl Acad Sci USA 94:1629–1633
123